AWS、Azure、GCP的生成式AI产品

ChatGPT, which is mainly invested by Microsoft, sparked a wave of generative AI. Corresponding google launched bard to fight and defend. The top three cloud vendors AWS, Azure, and GCP have also launched cloud services related to generative AI. This article summarizes the three corresponding main products.

ChatGPT引发了生成式AI的火热浪潮，ChatGPT主要是由微软进行投资的。对应的Google推出了bard以对抗和防守。前三的云厂商AWS、Azure、GCP也推出了生成式AI相关的云服务。本篇就总结下这三家对应的主要产品。

一、AWS Generative AI

AWS offer 4 layers machine learning service (AI services, AI platforms, AI frameworks, AI Infrastructure):

AWS Generative AI (AI) is a suite of services that use machine learning to create new content, such as images, text, and code. These services can be used for a variety of purposes, such as:

• Generating images: AWS Generative AI can be used to create realistic images of people, objects, and scenes. This can be used for a variety of purposes, such as creating marketing materials, generating training data for machine learning models, or simply creating art.
• Generating text: AWS Generative AI can be used to create realistic text, such as news articles, product descriptions, or even creative writing. This can be used for a variety of purposes, such as generating content for websites, creating marketing materials, or simply writing stories.
• Generating code: AWS Generative AI can be used to generate code, such as Python scripts, Java programs, or even C++ functions. This can be used for a variety of purposes, such as automating tasks, developing new software, or simply learning how to code.

AWS Generative AI is still under development, but it has the potential to revolutionize the way we create content. By automating the process of creating new content, AWS Generative AI can help us to be more productive and creative.

Here are some of the AWS Generative AI services:

• Amazon Polly: Amazon Polly is a service that can be used to generate human-like speech from text. This can be used to create audiobooks, generate synthetic voice for customer service chatbots, or simply create immersive experiences.
• Amazon Lex: Amazon Lex is a service that can be used to build conversational AI applications. These applications can be used to interact with customers in a natural way, such as through voice or text.
• Amazon Rekognition: Amazon Rekognition is a service that can be used to detect objects, faces, and text in images and videos. This can be used for a variety of purposes, such as identifying people in security footage, detecting product defects, or simply understanding the content of images and videos.
• Amazon SageMaker: Amazon SageMaker is a service that can be used to build, train, and deploy machine learning models. This can be used to create custom generative AI models that can be tailored to specific needs.
• CodeWhisperer: CodeWhisperer is an AI pair programming tool that offers developers an intuitive, secure, and highly personalized coding experience. The tool examines a developer’s code and comments, taking into account their coding style, variable names, and cursor location to generate personalized code snippets. This level of customization not only makes their work more accurate but also helps streamline the coding process, saving valuable time. CodeWhisperer addresses copyright concerns, which can be especially crucial in the age of AI-generated code. Security is also a top priority for CodeWhisperer. The tool scans code for potential security issues and vulnerabilities, drawing on Amazon’s vast experience with managing large codebases and working with CodeGuru.
• Amazon Bedrock: Simplifying Access to High-Performing Foundation Models. Amazon Bedrock is a new service that brings together FMs from leading companies like AI21 Labs, Anthropic, Stability AI, and Amazon itself. Offering access to a range of powerful models for text and images, Bedrock empowers developers with the tools they need to build cutting-edge AI applications without the headache of managing complex infrastructure.

Reference Pages:

AI Services on AWS
Amazon Generative AI and Software Development
Overview of all AI based Amazon Web Services (AWS)

二、Azure Generative AI

Azure Generative AI is a suite of services that use machine learning to create new content, such as images, text, and code. These services can be used for a variety of purposes, such as:

• Generating images: Azure Generative AI can be used to create realistic images of people, objects, and scenes. This can be used for a variety of purposes, such as creating marketing materials, generating training data for machine learning models, or simply creating art.
• Generating text: Azure Generative AI can be used to create realistic text, such as news articles, product descriptions, or even creative writing. This can be used for a variety of purposes, such as generating content for websites, creating marketing materials, or simply writing stories.
• Generating code: Azure Generative AI can be used to generate code, such as Python scripts, Java programs, or even C++ functions. This can be used for a variety of purposes, such as automating tasks, developing new software, or simply learning how to code.

Azure Generative AI is still under development, but it has the potential to revolutionize the way we create content. By automating the process of creating new content, Azure Generative AI can help us to be more productive and creative.

Here are some of the Azure Generative AI services:

• Azure Cognitive Services – Text-to-Speech: This service can be used to generate human-like speech from text. This can be used to create audiobooks, generate synthetic voice for customer service chatbots, or simply create immersive experiences.
• Azure Bot Service: This service can be used to build conversational AI applications. These applications can be used to interact with customers in a natural way, such as through voice or text.
• Azure Computer Vision (kinect-dk): This service can be used to detect objects, faces, and text in images and videos. This can be used for a variety of purposes, such as identifying people in security footage, detecting product defects, or simply understanding the content of images and videos.
• Azure Machine Learning: This service can be used to build, train, and deploy machine learning models. This can be used to create custom generative AI models that can be tailored to specific needs.

三、GCP Generative AI

Google Generative AI (AI) is a suite of services that use machine learning to create new content, such as images, text, and code. These services can be used for a variety of purposes, such as:

• Generating images: Google Generative AI can be used to create realistic images of people, objects, and scenes. This can be used for a variety of purposes, such as creating marketing materials, generating training data for machine learning models, or simply creating art.
• Generating text: Google Generative AI can be used to create realistic text, such as news articles, product descriptions, or even creative writing. This can be used for a variety of purposes, such as generating content for websites, creating marketing materials, or simply writing stories.
• Generating code: Google Generative AI can be used to generate code, such as Python scripts, Java programs, or even C++ functions. This can be used for a variety of purposes, such as automating tasks, developing new software, or simply learning how to code.

Google Generative AI is still under development, but it has the potential to revolutionize the way we create content. By automating the process of creating new content, Google Generative AI can help us to be more productive and creative.

Here are some of the Google Generative AI services:

• Google Cloud AutoML Vision Edge: This service can be used to train and deploy machine learning models on edge devices, such as smartphones and cameras. This can be used to create applications that can generate images, text, or code on-device, without having to send data to the cloud.
• Google Cloud AutoML Text-to-Text Transfer: This service can be used to transfer the knowledge of a pre-trained machine learning model to a new model. This can be used to create custom generative AI models that can be tailored to specific needs.
• Google Cloud AutoML Video Intelligence: This service can be used to extract insights from video content, such as identifying objects, faces, and text. This can be used for a variety of purposes, such as creating closed captioning for videos, detecting product defects, or simply understanding the content of videos.
• Google Cloud TPUs: These are custom-designed chips that are optimized for machine learning workloads. They can be used to train and deploy large-scale machine learning models, such as those used for generative AI.